Stochastic gradient boosting for predicting the maintenance effort of software‐intensive systems

Abstract

The maintenance of software-intensive systems (SISs) must be undertaken to correct faults, improve the design, implement enhancements, adapt programmes such that different hardware, software, system features, and telecommunications facilities can be used, as well as to migrate legacy software. A lack of planning has been identified as one explanation for late and over budget software projects. An activity of planning is effort prediction. The goal of this study is to propose the application of a stochastic gradient boosting (SGB) model for predicting the SIS maintenance effort. We compare the SGB prediction accuracy with those obtained with statistical regression, neural network, support vector regression, decision trees, and association rules. We trained and tested the models with five SIS data sets selected from the International Software Benchmarking Standards Group Release 11. The SGB prediction accuracy was statistically better than the mentioned five models in the two larger data sets. We can conclude that a SGB can be applied to predict the maintenance effort of SISs coded in languages of the third generation and developed on either mainframes or multi-platform. The predicted effort corresponds to the aggregate of efforts obtained from the project team, project management, and project administration.

Electronic Customer Relationship Management in action: an example in business

“…We prove in this paper the usefulness of computational intelligence algorithms in eCRM, for the automatic classification of Online Shoppers Purchasing Intention. With respect to other and diverse applications, there is an ocean of possibilities [98][99][100][101][102][103][104][105][106][107][108][109][110][111][112][113][114][115][116][117]. As future work, we want to apply computational intelligence algorithm to other tourism and business activities.…”

Effort prediction for the software project construction phase

“…In accordance with the TD, there are studies specifically applying effort prediction models for new 3,20 and maintenance projects. 22,46 The procedure followed for training and testing the prediction models is termed the validation method. The three most used have been holdout (38%), leave-one-out cross validation (LOOCV, 37%) and k-fold cross validation (19%).…”

“…A study published in 2020 identifies only 16 studies published between 1995 and 2018 on maintenance effort prediction. 22 Regarding testing effort prediction, I only identified two studies published from 1998 whose conclusions are based on statistical significance. 21,53 Regarding TD methodology, there are systematic review studies on specific TD methodology such as agile, 54 and on software development teams distributed around the world (i.e., global development).…”

“…19 The published systematic reviews on hundreds of studies about SDLC effort prediction have been built having as exclusion criterion those studies aimed at specific phases such as testing or maintenance. 3,6,7,20 A few studies have proposed models to predict the effort of phases such as testing 21 or maintenance 22 ; however, I did not find any study whose objective has been to propose a model to predict specifically the construction effort (CE). Figure 1 depicts the SDLC phases and those activities specifically related to the construction phase.…”

Software Enhancement Effort Prediction Using Machine-Learning Techniques: A Systematic Mapping Study

“…Presenting a model and metrics for the adaptive maintenance effort estimation/prediction S10 [32] Adaptive, corrective, perfective, preventive Implementing, evaluating and improving software maintenance effort prediction model based on expert judgment method S11 [33] Adaptive, corrective, perfective, preventive Proposing six models based on eight different indicators of evolution activity, their predictive power is examined and compared to that of two baseline models. S12 [51] Enhancement…”